Larval Dispersal Modeling Reveals Low Connectivity among National Marine Protected Areas in the Yellow and East China Seas

Abstract

Marine protected areas (MPAs) are vital for protecting biodiversity, maintaining ecosystem integrity, and tackling future climate change. The effectiveness of MPA networks relies on connectivity, yet connectivity assessments are often skipped in the planning process. Here we employed a multi-species biophysical model to examine the connectivity patterns formed among the 21 national MPAs in the Yellow and East China Seas. We simulated the potential larval dispersal of 14 oviparous species of five classes. Larvae of non-migratory species with pelagic larval duration (PLD) were assumed to be passive floating particles with no explicit vertical migration. A total of 217,000 particles were released according to spawning period, living depth, and species distribution, and they were assumed to move with currents during the PLD. Most larvae were dispersed around the MPAs (0-60 m isobaths) and consistent with the currents. Larval export increased with PLD and current velocity, but if PLD was too long, few larvae survived due to high daily mortality during pelagic dispersal. The overall connectivity pattern exhibited a north-to-south dispersal trend corresponding to coastal currents. Our results indicated that the national MPAs in the Yellow and East China Seas did not form a well-connected network and nearly 30% of them were isolated. These MPAs formed three distinct groups, one in the Yellow Sea ecoregion and two in the East China Sea ecoregion. Four MPAs (all in coastal Zhejiang) emerged as key nodes for ensuring multi-generational connectivity. Under the pressure of future climate change, high self-recruitment and low connectivity present significant challenges for building well-connected MPA networks. We suggest adding new protected areas as stepping stones for bioecological corridors. Focused protection of the Yellow Sea ecoregion could have a good effect on the southern part of the population recruitment downstream. Conservation management should be adjusted according to the life cycles and distributions of vulnerable species, as well as seasonal changes in coastal currents. This study provides a scientific basis for improving ecological connectivity and conservation effectiveness of MPAs in the Yellow and East China Seas.

Keywords: biophysical modeling; connectivity; larval dispersal; marine protected areas; network.

Figure 1

Distribution of 21 national MPAs and major currents in the Yellow and East China Seas. The blue points are the central points of MPAs. Arrowhead lines show the current directions. Dashed arrowhead lines are the possible directions of YSCC, YDW, and ZFCC in summer [30,31,32,33]. YSCC: the Yellow Sea Coastal Current; YDW: Yangtze Diluted Water; ZFCC: Zhejiang–Fujian Coastal Current; TWC: Taiwan Warm Current; KC: Kuroshio Current. MNR and SMPA in right table were marine nature reserves and special marine protected areas, respectively.

Figure 2

Modeling quarterly mean velocity and direction of ocean currents in Yellow and East China Seas: (a) January–March, (b) April–June, (c) July–September, and (d) October–December. The length of the line indicates the average velocity, and the arrows indicate the current direction.

Figure 3

Total larval abundance at the end of pelagic larval duration. The color indicates the density particles. Particles of all 14 species were superimposed and were given the same weight.

Figure 4

Dispersal fluxes of 14 species in each MPA. The species are arranged from top to bottom according to the length of PLD.

Figure 5

Schematic diagram of connections between MPAs and the network node importance metrics. Circle sizes indicate the relative sizes of protected areas. The directions of the arrows represent the direction of species particle dispersal. Dashed lines indicate from south to north, and solid lines indicate from north to south. Thickness of the lines represents the subsidized recruitment rate between MPAs. The network node importance metrics are shown in (a) Betweenness centrality, (b) Indegree centrality, (c) Outdegree centrality, and (d) Degree centrality.